The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute system commands including gh (GitHub CLI) and jq (JSON processor). These are used to query and modify GitHub Project board items and fields.
[INDIRECT_PROMPT_INJECTION]: The skill's core functions rely on processing data from external GitHub project boards, which represents a surface for indirect prompt injection.
Ingestion points: Project metadata is ingested via the GH_CACHE_ITEMS and GH_CACHE_FIELDS environment variables, which are populated from external GitHub API responses.
Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the ingested project data as untrusted or to ignore embedded instructions.
Capability inventory: The skill possesses the capability to perform network operations and state modifications on GitHub via the gh project CLI commands in functions like add_issue_to_project and set_project_status.
Sanitization: The skill employs shell variable quoting and uses jq to extract specific fields, which mitigates direct shell injection but does not prevent malicious content in the data from influencing the agent's logic or downstream responses.

project-board-enforcement